mirror of
https://github.com/ROCm/composable_kernel.git
synced 2026-05-14 10:09:41 +00:00
ff1d9f88fa
* Update license year
* Add initial code to override decode problem
* Fix splitkv traits/args overriding error
* Reshape and transpose lse for decode
* Remove debug code
* Prettify example code
* Use better function name
* Add kMergeNumHeadGroupsSeqLenQ flag
Kernel user can use this switch to turn on/off optimization for
some problem sizes
* Add missing flag declarations
* Default turn off kMergeNumHeadGroupsSeqLenQ in codegen
* Group similar statements together
* Remove assumption of seqlen_q=1
* Remove kMergeNumHeadGroupsSeqLenQ from splitkv combine kernel
* Support kMergeNumHeadGroupsSeqLenQ=true in fmha splitkv kernel
* Run kMergeNumHeadGroupsSeqLenQ=true kernels when need
* Fix group mode block skip logics
* Undo changes of normal fwd kernel
* Update in GridSize() and using GridSize() for splitkv kernel (#1799)
---------
Co-authored-by: Qianfeng <qianfeng.zhang@amd.com>
[ROCm/composable_kernel commit: 24b12d04af]
37 lines
1.7 KiB
C++
37 lines
1.7 KiB
C++
// SPDX-License-Identifier: MIT
|
|
// Copyright (c) 2018-2025, Advanced Micro Devices, Inc. All rights reserved.
|
|
|
|
#pragma once
|
|
|
|
#include "ck_tile/host/arg_parser.hpp"
|
|
#include "ck_tile/host/check_err.hpp"
|
|
#include "ck_tile/host/convolution_host_tensor_descriptor_helper.hpp"
|
|
#include "ck_tile/host/convolution_parameter.hpp"
|
|
#include "ck_tile/host/device_memory.hpp"
|
|
#include "ck_tile/host/fill.hpp"
|
|
#include "ck_tile/host/hip_check_error.hpp"
|
|
#include "ck_tile/host/host_tensor.hpp"
|
|
#include "ck_tile/host/joinable_thread.hpp"
|
|
#include "ck_tile/host/kernel_launch.hpp"
|
|
#include "ck_tile/host/ranges.hpp"
|
|
#include "ck_tile/host/reference/reference_batched_dropout.hpp"
|
|
#include "ck_tile/host/reference/reference_batched_elementwise.hpp"
|
|
#include "ck_tile/host/reference/reference_batched_gemm.hpp"
|
|
#include "ck_tile/host/reference/reference_batched_masking.hpp"
|
|
#include "ck_tile/host/reference/reference_batched_rotary_position_embedding.hpp"
|
|
#include "ck_tile/host/reference/reference_batched_softmax.hpp"
|
|
#include "ck_tile/host/reference/reference_elementwise.hpp"
|
|
#include "ck_tile/host/reference/reference_fused_moe.hpp"
|
|
#include "ck_tile/host/reference/reference_gemm.hpp"
|
|
#include "ck_tile/host/reference/reference_im2col.hpp"
|
|
#include "ck_tile/host/reference/reference_layernorm2d_fwd.hpp"
|
|
#include "ck_tile/host/reference/reference_moe_sorting.hpp"
|
|
#include "ck_tile/host/reference/reference_permute.hpp"
|
|
#include "ck_tile/host/reference/reference_reduce.hpp"
|
|
#include "ck_tile/host/reference/reference_rmsnorm2d_fwd.hpp"
|
|
#include "ck_tile/host/reference/reference_rowwise_quantization2d.hpp"
|
|
#include "ck_tile/host/reference/reference_softmax.hpp"
|
|
#include "ck_tile/host/reference/reference_topk.hpp"
|
|
#include "ck_tile/host/stream_config.hpp"
|
|
#include "ck_tile/host/timer.hpp"
|